The Phillips Curve: Read the Fine Print
The Federal Reserve’s Dual Mandate is the Phillips Curve.
It’s not that the Fed is charged with the impossible task of reaching a state of 2% inflation and full employment; rather, the Fed is trying to obtain these two goals, believing (according to the Phillips Curve), these goals come as a trade-off needing to be managed.
Just two weeks ago, Chair Powell spoke of this on-going battle, like a never-ending tug of war between inflation and employment, as reported by Reuters:
We have to balance those two goals when they are in tension, as they are right now.
This tension, as described by Powell, is not just another Powell-ism, or something the head of America’s central bank said off the cuff in an interview. Instead, this idea of balancing inflation against employment has been a widely discussed phenomena in academia and by central bankers for several generations. The current CEO of the Federal Reserve Bank of New York, a man who may be more powerful than Powell himself, co-authored a paper in 2014 explaining the trade-off, calling it the “divine coincidence of monetary policy,” whereby:
…the central bank can simultaneously stabilize both employment and inflation by altering a single policy instrument—the short-term nominal interest rate.
According to one of the current members of the inner circle at the Fed, just by changing interest rates, employment and inflation will simultaneously stabilize.
The above quotes are only two examples. But countless instances exist where economic planners discuss the necessity of balancing the two objectives, as if it’s a zero-sum game.
To best illustrate the juggling act, the St. Louis Federal Reserve Bank explains the Phillips Curve, or the belief there is a trade-off between inflation and unemployment:
According to economic planners, human behavior can be plotted on a simple chart. The higher the inflation, the more people will work. The lower the inflation, the less people will work. If the theory is sound, then a country like Turkey, with an annual inflation rate of 21.3% should be struggling with employment that is too high, instead of struggling with bread lines.
Unfortunately, Curve supporters may retort, with the highly untenable position, that the model doesn’t work in some countries, like it works here….
But there is always the fine print to consider. In the model presented by the St. Louis Fed above, a tiny asterisk at the bottom of the chart reads:
This illustration is intended for conceptual purposes only. It's partly modeled on Figure 1 scatterplot on Page 285 of Phillips’ 1958 paper, which contained 1861-1913 data. Each dot represents a year. The vertical axis shows the average rate of change of money wage rates; the horizontal shows the average unemployment.
This is Figure 1 from the original paper:
Phillips’ paper became the cornerstone of monetary policy for the entire world. Central bankers to this day look to balance both inflation against unemployment in harmonious equilibrium. Written in 1958, he used data from the United Kingdom that pre-dated the abolition of the slavery! However, it’s time to acknowledge that these ideas have long since outlived their usefulness and are now used to conduct economic harm above all else.
For argument's sake, even if the UK data from 1861 was reliable, showing a correlation between wage rates and unemployment, and even if the correlation equaled causation and all assumptions used in the paper (like an “increase in productivity of 2 per cent per year,” or after “Ignoring years in which import prices rise rapidly enough to initiate a wage-price spiral”), were all considered true, the question becomes: Why would this be relevant to America today?
The problem of measurement uncertainty and errors are one thing. But even if the data was adequately measured, there’s no plausible reason to think this has any sort of predictability or relevant application in the future. If the relationship existed for the last 50 years, there’s no reason the relationship should exist next year.
To make matters worse, looking at the last 50 years of inflation vs. unemployment is more than sufficient to see no such relationship exists. The St. Louis Fed, calls it the Cloud of Points:
For all the talk of quantitative economists and the use of data as an economic tool, apparently no one at the Fed has ever run a correlation coefficient or R-Square to see if there still is, or ever was, a relationship between the two.
But here we are, sadly. And 2022 is starting to look interesting. America is experiencing the highest rates of inflation in several decades. The expansion of the money supply, national debt and the Fed’s balance sheet continue to garner little to no attention. At some point in the not too distant future, should Chair Powell announce they are raising rates to fight inflation, he’ll cite how it worked in the past and claim the data supports the decision. While no one likes being lied to, at least it feels better to know you are being lied to, as opposed to thinking any of their policies are backed by credible economic theory.